Xerox originally developed the adaptive controls to manage complicated printing tasks, such as the robotics involved in moving paper at 100 pages per minute.
“It is very complicated, because you need to know where the paper is and how to control it,” said Mark Bernstein, president of Xerox's Palo Alto Research Center.
To handle such tasks, the company came up with some model-based controls, in which different parts know their capabilities and communicate them back to a control module so that parts can be added or exchanged, said Scott Elrod, manager of the center’s Clean Technology Program.
Using the same reasoning, PARC has developed software that can reduce servers’ energy usage by 30 percent (or, more likely, allow data centers to provide 30 percent more service using the same energy), he said.
The software basically predicts demand, allowing data centers to prioritize and manage jobs more efficiently, he said.
Similar control software could be used to monitor and control electricity demand on the grid or in buildings, he said, but the first application is likely to be increasing data-center efficiency.
PARC is starting to talk to some companies with large data centers “right now,” Elrod said.
Because of the many different data-center architectures, the center expects to commercialize its technology using a client-service collaboration, in which PARC will modify and install the technology in data centers for its clients, Bernstein said.